Distributed Fair Valuation: Alternative View on Fair Value
Accounting from a Distributed Cognition Perspective1
Noriaki Okamoto (nokamoto@rku.ac.jp)
Last revised: May 2012
Abstract
This exploratory study considers controversial fair value accounting from a unique
interdisciplinary perspective. Given the growing influence of esoteric financial instruments
whose valuation is becoming increasingly complex and difficult, it is necessary to turn our
attention to investment companies’ valuation practices. This paper considers some actual
conditions of valuation practices and suggests that the processes of valuation of complex
financial instruments can be distributed among individuals who are located inside and/or
outside of an investment company. This point is aptly explained from the viewpoint of
distributed cognition, which focuses on reporting companies’ external and internal
resources for fair value accounting. It thus facilitates a theoretical understanding of
increasingly specialized, complicated, and professionalized accounting valuation.
Furthermore, the results of some experimental cognitive studies imply that interactions of
distributed knowledge among human beings may contribute to fairer valuations or
objective accounting numbers.
1
The author gratefully acknowledges helpful comments and suggestions from David Hatherly
and Donald MacKenzie.
1
1.
Introduction
After the revelation of the devastating effects of the global financial crisis, fair value
accounting (FVA) has been under severe attack by politicians 2 and regulatory authorities 3,
although it has been studied by many accounting scholars for several decades. Recently,
suspiciously labeling FVA as one of the causes of massive damages of the financial crisis
has rekindled the debate over what appropriate accounting valuation is.4
However, as some have cast doubts on the significance of FVA, the appropriateness
of FVA is still questionable.5 Proponents as well as opponents of FVA have occasionally
criticized arguments of their counterparts6 as both of them have derived their conclusions
on the basis of specific consequences of FVA.7 Although previous literature has indicated
various consequences of FVA 8 , we still do not have clear criteria for judging the
appropriateness of FVA. However, the fact that the International Accounting Standards
Board (IASB) recently issued a standard on FVA (IFRS 13) implies further global diffusion
and penetration of FVA.
The question then arises as to why accounting standard setters such as the IASB and
FASB have pursued FAV. To address this question, the process of accounting standard
setting is initially considered. Assuming that accounting standard setting depends on the
standard setter’s assessment of similarities and differences of corporate transactions, the
critical process “cannot be regarded as value-free judgments resulting from a strictly
technical process” (Young and Williams 2010, p. 511). Thus, accounting standard setters
2
For example, French President Jacques Chirac stated in 2003 that the IAS would have to
face “nefarious consequences for financial stability” (Galligan 2007, p. 276).
3 For a recent review of the United States and the FASB (Financial Accounting Standards
Board)’s actions, for example, see Bougen and Young (2012).
4 The impact of FVA on the recent financial crisis tends to be discussed in terms of
pro-cyclicality. For a discussion of the matter, for example, see IMF (2008, Chapter 3). This
paper does not examine the economic consequences of pro-cyclicality of FVA.
5 For example, if the word “financial accounting” was chosen as a search term (in Title,
Abstract ID & Keywords with All Dates) and related papers were searched in the social science
search engine, Social Science Research Network (SSRN: http://www.ssrn.com), 615 papers
were hit (accessed 6 January 2012). Then, if the same thing was done for the term “fair value
accounting”, 214 papers were hit. This simply means that more than one third of total financial
accounting papers would be equivalent to FVA-related papers.
6 Bignon et al. (2009) exceptionally take a balanced as well as neutral argument on FVA.
7 For one of the most comprehensive reviews of FVA research, see Obinata (2011).
8 As to the summary of pros and cons of FVA, for example, see Penman (2008, Appendix B)
and Ryan (2008).
2
might not fully understand the clear advantages of FVA over other valuation models
because their political decisions have been based on subjective value-judgments rather
than objective evidences. Based on this assumption, it is still uncertain whether FVA really
reflects the fair value of corporate assets/liabilities and is superior to traditionally-accepted
historical cost accounting (HCA). Even in one of the top-ranked accounting journals which
has favored economics/statistics-based research, Kaplan (2011, p. 377) warns that “all this
[accounting] scholarship takes the “fair value” as given and does not explore how fair
values actually get estimated. The accounting profession is in danger of potentially
outsourcing this critical measurement, and its valuation, to others.” This provocative
remark suggests the importance of investigating practical as well as pragmatic aspects of
FVA.
With this view, the present study focuses on some aspects of FVA practice and takes
the concept of distributed cognition as an interpretative viewpoint. The consideration below
indicates that the practice of FVA of complex financial instruments is to some extent
distributed. It also reveals that the practice of FVA of those financial instruments
incorporates different tools and knowledge, and that aspect could make the practice be
perspectivistic. To take the analysis further, the fairness of FVA is explored based on some
cognitive studies. This is fairly important since current analytical views of FVA have
originated mainly from economics. Therefore, the aim of this paper is to present an
interdisciplinary perspective on FVA by drawing attention to the intersection between FVA
and the perspective of distributed cognition, and to provide a different understand ing of
FVA. Even though the perspective may be interpretative and heterodox, “the more ways we
have of thinking about and acting in the world the better” (Pickering 2009, p. 204).
In the next section, this paper reviews the controversy over FVA. Then, based on a
brief assessment of the development of FVA, it specifies the scope of this paper. In the
third section, it examines information regarding the actual practice of FVA. Considering the
recent global financial crisis and the controversy over FVA, the particular focus is on FVA of
complex financial instruments in investment companies. Based on an analysis of FVA
practice in the area where valuation seems to be relatively difficult, this study emphasizes
that knowledge of actors involved in FVA can be distributed inside and outside reporting
3
companies. In the fourth section, the author considers the concept of distributed cognition
which gets attention from a wide variety of disciplines, and it gives a different
understanding of FVA. The fifth section discusses the main findings of this paper and
identifies future research.
2.
FVA: Panacea or Distress for Financial Accounting?
2.1 The Controversy over FVA in the Context of Accounting Regulation
FVA has been applied to various corporate assets/liabilities during the development of
major accounting standards. The US standard of SFAS (Statement of Financial Accounting
Standards) 157 defines “fair value” as “the price that would be received to sell an asset or
paid to transfer a liability in an orderly transaction between market participants at the
measurement date” (FASB 2008, para. 5). The IASB recently issued “IFRS 13-Fair Value
Measurement“, and the requirements of both standards are quite similar.
FVA is markedly different from traditional HCA. “A fundamental conceptual issue
[facing accounting standard setters] is the extent to which the standard should move away
from traditional cost based accounting to marking assets and liabilities to market,
euphemistically referred to as ‘fair value’ accounting. 9” Typically, proponents of FVA have
criticized HCA and emphasized the relative advantages of FVA, such as better reflection of
corporate reality or risk management activity. They look at the HCA-based balance sheet
and call it a disgrace: “Reporting assets at their historical cost is like driving down the road
looking in the rear-vision mirror!” (Penman 2010, p. 169). Such worship of FVA is regarded
as “the fair value paradigm (Hitz 2007)” or “the full FVA ideology (Georgiou and Jack 2011)”.
“The fair value paradigm rests on the decision usefulness paradigm, which was established
as an official standard setting objective only with the formation of the FASB and the
conceptual framework project” (Hitz 2007, p. 327).
The empirical literature on FVA, especially before the financial crisis, claimed “the
relevance of fair value measurement can only be supported for securities traded on highly
9
Speech by Paul A. Volcker, Chairman of the Trustees, of International Accounting Standards
Committee Foundation in a statement before the Capital Markets, Insurance and Government
Sponsored Enterprises Subcommittee of the U.S. House of Representatives, Washington DC,
June 7, 2001 (Penman 2010, p. 167).
4
liquid markets, while the evidence reinforces the significance of the reliability objection
both for financial and non-financial assets (Hitz 2007, p. 325)”. The critical core of FVA is
that “Fair value accounting finds its justification in efficient market theory; price
summarizes all available information, so price supplies the accounting for value” (Penman
2010, p. 170). Furthermore, some empirical research has documented correlations
between fair value measurements and stock prices that are useful for understanding
whether fair values are “relevant to investors”. Unfortunately, however, it does not provide
much guidance on the policy question of whether fair values should be reported in place of
HCA (Penman 2008, p. 7).
Thus, modern accounting regulation has so far maintained the “mixed-attribute
accounting model” which mixes two types of measurement models (HCA and FVA) in a set
of accounting standards.10 The mixed-attribute model often allows firms to choose the
measurement attribute they desire for a position through how they classify the position
(Ryan 2008, p. 7).11 Accordingly, under current major accounting standards, unrealized
gains and losses can be recorded in accumulated other comprehensive income (or a
component of owners’ equity) or in net income. In that sense, the classification of
gains/losses arising from FVA is also said to be mixed.
From a regulatory perspective, whether it is appropriate to sustain the mixed valuation
model for future accounting regulation is still controversial. On the one hand, it is intuitively
simple and straightforward to adopt either of the two valuation methods. On the other hand,
it may be better to sustain the current “mixed-attribute accounting model” because each
valuation model seems to have its own strengths and weaknesses. Before considering this
tough question, it would be useful to outline how FVA has expanded its sphere of influence
in modern accounting regulation.
2.2 A Review of the Historical Trajectory of FVA
There are only a limited number of studies that have comprehensively explained how
10
To put it another way, FVA is not yet fully institutionalized as an accounting practice
(Georgiou and Jack 2011, p. 313)
11 In other words, fair value measurement guidance has been primarily contained in a
cross-referenced patchwork of accounting standards related to financial instruments (AAA
Financial Accounting Standards Committee 2005, p. 187).
5
FVA has been adopted in major accounting standards such as the IFRS or the US FASB
standards. Notably, Georgiou and Jack (2011) attempt to highlight the historical
development of FAV in the UK and US using three time periods, 1930―1970, 1970―1990,
and 1990―the present.12 In the first period, which began after the Wall Street Crash and
included the subsequent depression in the US in the 1930s , reporting by UK and US
companies was characterized by the concept of ‘reflecting the business’, and prescriptive
formats and methods imposed by regulation were resisted (Georgiou and Jack 2011, p.
314). As a result, corporate uses of mixed measurement incorporating market values
maintained a degree of legitimacy during that period. After the crisis in the late 1930s in the
US, the debate over HCA versus current value heated up. However, during the period 1940
―70, the debate continued with arguments from both academics and practitioners about
the lack of appropriate fair representation of business activity in cost accounting showing
that full cognitive acceptance of the basis was never achieved in that time (Georgiou and
Jack 2011, p. 316).
In the second period, between 1970―1990, the Saving and Loans crisis (commonly
referred to as the S&L crisis) was a significant event that affected the displacement of HCA
in the US. The crisis exposed the deficiency of HCA and the Securities Exchange
Commission (SEC) advised the FASB to advocate for market values for certain assets
(Georgiou and Jack 2011, p. 317). Furthermore, market developments since the mid-1980s,
especially the ease in determining fair values of financial instruments and the increased
use of off-balance sheet financing, resulted in the issuing of standards which promoted
disclosure of fair value, such as SFAS 105 and SFAS 107 (Brown-Hruska and Satwah
2009, p. 137). Subsequently, SFAS 115 Accounting for Certain Investments in Debt and
Equity Securities was issued, and this was crucial in the regulatory move from HCA-based
accounting to market value-based accounting.
In the third period, since 1990, more and more HCA-based accounting standards have
been replaced by FVA-based ones. For example, not only SFAS 157 in the US, but also
FRS (Financial Accounting Standards) 7 Fair Values in Acquisition Accounting in the UK
12
Since the purpose of this study is not to review a long history of the two distinctive
accounting valuation models, it mainly focuses on Georgiou and Jack (2011) and roughly
depicts how FVA came to be accepted within the regulatory framework.
6
and IASC’s IAS 39 Financial Instruments: Recognition and Measurement are typical
example of those standards (Georgiou and Jack, 2011, p. 319). Recent movement towards
“Mark-to-market might be implemented not to solve a problem in private markets, but to
remove regulatory discretion provided by historical-cost accounting. That rule, however,
may have gone too far by eliminating efficient forbearance by market participants who,
after all, were not fooled by historical-cost accounting the first time round” (Epstein and
Henderson 2011, p. 521).
Based on this brief summary of the development of FVA, it can be inferred that neither
FVA nor similar market value-based valuation nor HCA has fully dominated previous
accounting practice. Rather, the use of mixed-measurement has been accepted for a long
time. Nevertheless, under the current accounting regulation by the IASB and FASB which
exert the dominant global influence, the objects subject to FVA are diverse and highly
complex. Given that the economic and business conditions are different from earlier times,
it is unclear whether the mixed valuation model is able to cope with those complex financial
instruments in a systematic way. With respect to this point, Koga (2001) argued that in the
era of a relatively finance-oriented economy, or what is called the ‘financialized 13’ world,
financial and intangible assets/liabilities are important and the reflection of their volatility
and corporate risk management activities in financial accounting are essential. In that case,
finance-oriented accounting regulation which is underpinned by FVA can be more
appropriate. This is in contrast to previous HCA-based accounting regulation which
seemed to have been appropriate for the production-oriented economy where
manufacturing/producing companies were comparatively dominant in the economy.
This assertion implies that at this moment it is more important than ever to pay
attention to the peculiarities of current FVA practice in financialized corporate activities. 14
According to Epstein (2005, p. 3), “financialization means the increasing role of financial
motives, financial markets, financial actors and financial institutions in the operation of the
domestic and international economies.”
14 The background might be caused by an increase in the proportion of the income generated
by the industrial/post-industrial economies which accrues to those engaged in the finance
industry, as a consequence of three things: the growth in and increasing complexity of
intermediating activities, very largely of a speculative kind, between savers and the users of
capital in the real economy; The increasingly strident assertion of the property rights of owners
as transcending all other forms of social accountability for business corporations; Increasing
efforts on the part of government to promote an “equity culture” in the belief that it will enhance
13
7
An investigation of the pragmatic aspect of FVA for financial instruments can fill the gap
between practitioners coming to grips with FVA and accounting standard setters (or
academic accounting researchers). This type of awareness has been seen in social
studies of highly awkward complex finance and accounting practices.15
3.
The Detailed Analysis of FVA Practices
3.1 Current Standards of FVA
Before investigating the actual practices of FVA, it is necessary to summarize the
latest requirements for FVA in formal accounting standards. Although the development of
FVA has been tortuous, it has resulted in the establishment of SFAS 157, an independent
FVA standard in the US. Figure 1 below depicts the calculation of the fair value of corporate
assets based on SFAS 157. The standard was welcomed because a single standard would
increase the efficiency and consistency of measuring fair value across the many standards
that require fair value reporting and disclosure (AAA Financial Accounting Standards
Committee 2005, p. 188). It distinguishes cases where market prices for identical assets or
liabilities are readily available from active markets (so-called Level 1 measurement) and
cases where hypothetical market prices have to be estimated (Level 2 and Level 3,
differentiated by an increasing degree of subjectivity in making estimates). Also, as shown
in Figure 1, SFAS 157 places heavy emphasis on the inputs in measurement of assets and
liabilities.
SFAS 157 was set by the FASB and has been effective for corporate financial
statements issued in the US since November 2007. 16 Although the IASB issued a
comparable standard, IFRS 13, in May 2011, this paper relies on the US standard because
the ability of its own nationals to compete internationally (Dore 2008, p. 1098).
15 For instance, see Millo and MacKenzie (2009), MacKenzie (2009) and Leung (2011).
16 The FASB issued valuation guidance in 2009 called FASB Staff Position Financial
Accounting Standards (FAS) 157-4, “Determining Fair Value When the Volume and Level of
Activity for the Asset or Liability Have Significantly Decreased and Identifying Transactions
That Are Not Orderly”. The guidance, which is now codified in Accounting Standards
Codification (ASC) 820, highlighted considerations for valuing assets and liabilities in a difficult
market environment and clarified certain disclosure requirements. In early 2 010, the FASB
provided additional disclosure guidance in Accounting Standards Update No. 2010-06—“Fair
Value Measurements and Disclosures (Topic 820): Improving Disclosures about Fair Value
Measurements”.
8
there is more evidence of its application.
Is it possible to get
observable market
inputs?
YES
S
Does it reflect
quoted price?
NO
Corroborated by
other market data?
NO
YES
S
Is it for identical asset
or liability?
YES
YES
NO
S
Is it in active
market?
NO
S
NO
YES
S
LEVEL 1
LEVEL 2
LEVEL 3
Inputs reflect
quoted prices in an
active market
Indirect inputs are
derived from similar
asset or liability
Inputs are derived
through extrapolation or
interpolation
Figure 1 Fair Value Hierarchy in SFAS 157
This figure is based on Figure 1 of Campbell et al. (2008).
The role of auditors for FVA is also important. External auditors may alter the reporting
company’s FVA figures in the auditing process. Most auditors, however, have little training
in valuation (Martin et al. 2006, p. 289). Further, “no amount of auditing can remove the
underlying
estimation
uncertainty
of
reported
values
that
are
determined
by
management-derived estimation models that are hypersensitive to small changes in
9
inputs—inputs that are subjectively chosen by management from within a reasonable
range (Christensen et al. 2012, p. 128).” In addition, auditors prefer to follow appropriate
guidance on FVA auditing (Fitzsimons et al. 2010).
Even so, International Standard on Auditing 500: Audit Evidence, published by the
International Federation of Accountants, states that auditors may accept the findings of a
specialist hired by management as appropriate audit evidence. This is not a natural
process arising through the division of multiple-disciplinary labor; it may signify specific
and often competing hierarchical relations between different bodies of expertise (Power
1997, p. 140). Thus, “the specialization of tasks reinforces the auditor’s role as conductor
of an orchestra, and the need to acquire the necessary competence in valuation methods
to be able to ensure there is good coordination between specialists concerned
(Jacquemard, 2007, p. 279).” Therefore, it is useful to consider FVA from the perspective of
differently distributed cognition.
3.2 FVA of Complex Financial Instruments in Investment Companies
3.2.1 Types of Funds
Hereafter, the central focus is on FVA for financial instruments. Even though financial
instruments themselves are varied and have their different natures, most of them are
currently required to be valued at fair value. In this subsection, the valuation of portfolio
securities of investment companies (occasionally classified as funds) will be examined. It is
obvious that the problematic aspect of the SFAS157 is the Level 2 and 3 valuations. It is
unclear whether much attention is being given to accounting standards for external
reporting by private investment companies compared with other listed financial institutions
such as banks. However, the fact that accounting firms (such as KPMG (2011) and PWC
(2011)) provide services to support financial reporting by private investment companies
shows the growing significance of accountable calculation to those companies particularly
after the recent financial crisis.
There are several types of funds worldwide. According to TheCityUK (2011), there are
generally three types of funds: conventional, alternative, and private wealth funds.
Conventional funds include pension, mutual, and insurance funds, and the amount of
10
managed assets in this category at the end of 2010 was $79.3 trillion.17 Alternative funds
include hedge, private equity, exchange traded, and sovereign wealth funds, and managed
assets in this category at the end of 2010 were $10 trillion (TheCityUK 2011). Private
wealth funds manage in total $42.7 trillion of assets (around one-third of this amount is
incorporated in conventional investment management).
Each type of funds may have its own culture. For example, some funds have close
connections with affiliated banks and even work as their in-house department. However,
the purpose of this study is not to analyze practices in a specific type of funds, but to find a
unique common aspect of FVA in diverse investment companies. Information for this
research come from different sources. They consist of analysis of valuation practices in the
funds of US leading investment banks at the beginning of the recent crisis, journalists’
detailed ethnographic case studies of the valuation of exotic financial instruments in the US
and France, and the results of eight consecutive investigations of asset managers’
accounting practices. Assuming that it is difficult to get into the fiercely competitive world of
trading financial instruments, those research materials provide insightful characteristics of
FAV practice. With the exception of one French bank, the geographical context of those
materials is limited to the US as the country has a long history of regulating investment
companies. 18 For instance, in 1969, the US SEC issued guidance on fair valuation
determination in the context of restricted securities in Accounting Series Releases No. 113
(ASR 113) (October 21, 1969). 19 As a result of the introduction of SFAS 157, the board of
each investment company also needs to take into account the framework for measuring fair
17
The US remained by far the biggest source of the funds in 2010, accounting for nearly a half
of conventional assets under management or some $36 trillion (TheCityUK 2011).
18 The information from the case of French bank (Lépinay 2011, pp. 196-203) does not fully
indicate what type of accounting standards the bank used.
19 That release was followed by ASR 118 (December 23, 1970), in which the SEC sought to
provide more general valuation guidelines. In 1999, the staff of the Division of Investment
Management (the Staff) issued an interpretative letter to the Investment Company Institute (the
1999 Letter) (December 8, 1999) in an effort to clarify and provide additional guidance on
pricing issues in the case of an emergency or other unusual situations, and issued follow-up
advice in 2001 with regard to the valuation of foreign securities in the form of a second
interpretative letter to the Investment Company Institute (the 2001 Letter) (April 30, 2001).
More recently, the Staff provided “best practices” guidance in a speech by Lori Richards,
Director of the Office of Compliance, Inspections and Examinations (the Richards Speech)
(June 14, 2002). Each of these interpretations expanded the factors that investment company
boards of directors must take into account when making fair value determinations (Ake and
Hays 2007).
11
value in generally accepted accounting principles. In addition to accounting treatments,
even though those standards require greater quantitative and qualitative disclosure of the
details involved in pricing complex and narrowly-traded products, this paper mainly
focuses on the valuation process itself rather than the disclosure of detailed information
about FVA.
3.2.2 Distributed FVA: Outside an Organization
In the modern financial world, investment companies seek to employ skillful traders
who strategically place large amounts of other investors’ money into diversified complex
financial instruments to make trading profits. In terms of accounting for trading portfolios,
measurement of ambiguous level-2 and 3 items are particularly controversial. For example,
modern derivative contracts make the measurement of financial instruments complex. “The
virtual character of a derivative contract is enhanced if, as is increasingly the case, it can
be settled only by the transfer of cash, with neither party able to demand or impose delivery
of an underlying asset. The measure used to determine the amounts to be paid must
therefore be a ‘fact’: it must be an acceptable representation of the reality of which it
speaks, and not be subject to manipulation (MacKenzie 2009, p. 66).” The factual value of
complex financial instruments is expressed as follows.
Nothing in the financial world happens in a vacuum these days, given the exponential
growth of trillions of dollars of securities tied to the value of other securities―known as
“derivatives”―and the extraordinary complex and internecine web of global trading
relationships. Accounting rules in the industry promote these interrelationships by
requiring firms to check constantly with one another about the value of securities on
their balance sheets to make sure that value is reflected as accurately as possible.
Naturally, since judgment is involved, especially with ever more complex securities,
disagreements among traders about values are common (Cohan 2011, p. 4).
This quotation indicates that disagreements among sophisticated traders over values of
some financial instruments are frequent. The International Accounting Standards Board
Advisory Panel (2008) also accepts that professional valuations are often divergent at the
organizational level as follows:
12
As a result of applying judgment, two entities might arrive at different estimates of
the fair value of the same instrument even though both still meet the objective of fair
value measurement. This could be the case when, even if the two entities use the
same model, the unobservable inputs used in the model are different (para. 26).
Then, a matter of concern is how individuals in those organizations get the fair value of the
financial instruments in their portfolio. Do traders who are actively trading their assets
decide the fair value of traded assets? Or, do the members of the board who are generally
in charge of the traders’ business decide? Or, does a back office decide?
With regard to the valuation of assets, especially those in investment companies, Net
Asset Value (NAV) is a key term. It represents the value per share of the funds effectively
held by the fund computed at the end of each trading day. 20 “The fund’s NAV is used to
process purchases, redemptions, and exchange by shareholders. Proper valuation of
securities ensures that all transacting fund shareholders pay or receive a price that
represents their proportionate share of the fund’s portfolio (Investment Company Institute
(ICI) et al. 2006).” However, while the board is ultimately responsible for the fair valuation
process, that responsibility does not necessarily mean that the board itself must make fair
value determinations (ICI et al. 2006). According to the guidance of the SEC, a board may
direct members of fund management (who may or may not sit on the board) or others to
make the actual fair value determinations so long as the board reviews and approves the
methodology or methodologies by which fair value determinations are made, regularly
reviews the appropriateness and accuracy of the valuation methodologies, and makes any
necessary adjustments (ICI et al. 2006). Based on this information, senior board members
in those organizations may not need to be directly involved in FVA.
The quotation below shows how traders in a hedge fund actually valued complex
financial instruments whose fair market prices were not easily obtained. According to SFAS
157, it is possible to get an instrument’s fair value as an average of others’ estimates of
similar assets (FASB 2008, para. C91 21).
20
Other than accounting regulation, the Investment Companies Act of 1940 in the Unites
States provides that securities for which market quotations are “readily available” must be
valued at “fair value” as determined in good faith by a fund’s board of directors (ICI et al. 2006,
p. 1).
21 Bid-ask spread pricing methods appropriate under ASR 118 are also appropriate under
SFAS 157.
13
The way hedge funds-such as those run by Cioffi and Tannin 22―are required by the
Securities and Exchange Commission to value the securities they own is pretty
arcane.23 But it is based on the idea of taking an average of the prices other Wall
Street firms and other traders are finding in the market for similar securities, most
of which are thinly traded from one firm to another and are rarely traded by retail
investors or on exchanges, as with stocks. With these illiquid securities,
hedge-fund managers had to wait until the end of each month to get the marks from
other brokers and dealers, and then average them, and then report the “net asset
value,” or NAV as it is known on Wall Street, to investors (Cohan 2011, p. 548).
The paragraph below also reveals how FVA was being done at the beginning of the
financial crisis.
A week later, “knowing full well we’ve published our NAV”, according to this
executive, Goldman Sachs sent, by e-mail, its April marks on the securities to Cioffi.
“Now there’s a funny little procedure that the SEC imposes on you, which is that
even if you get a late mark, you have to consider it,” he said. “Suddenly we get
these marks. Except these marks are not marks from ninety-eight to ninety-seven.
They go from ninety-eight to fifty and sixty. Okay? You get it? They give us these
fifty and sixty prices. What we got from the other counterparties is ninety-eight. The
SEC rules say that when you do this, you either have to average them―but they’re
meant to be averaging ninety-sevens and ninety-eights, not fifties and ninety-eights
―or you can go and ask if those are the correct marks (Cohan 2011, p. 549).
In the above situation, how and why did Goldman Sachs (Goldman) alone give such low
marks at that moment? At that time, it is likely that Goldman expected the illiquid
mortgage market to continue to decline, and it was also selling and shorting those
securities (The Financial Crisis Inquiry Commission 2011, p. 236). Thus “Goldman has
been criticized―and sued―for selling its subprime mortgage securities to clients while
simultaneously betting against those securities”. A structured finance expert…reportedly
called Goldman’s practice “the most cynical use of credit information that I have ever
seen” and compared it to “buying fire insurance on someone else’s house then
committing arson” (The Financial Crisis Inquiry Commission 2011, p. 236). Here, the
22
They are former Bear Stearns hedge fund managers who managed sub-prime-laden hedge
funds (the Bear Stearns High Grade Structured Credit Strategies Master Funds Ltd and the
Bear Stearns High Grade Structured Credit Strategies Enhanced Master Funds Ltd).
23 According to the indictment, Bear Stearns Asset Management had an independent pricing
committee that was a group of professionals who were charged with overseeing the ultimate
calculation of the funds’ NAV. However, it was also bizarre that Cioffi could ask the committee
to give him his desired number.
(http://fl1.findlaw.com/news.findlaw.com/hdocs/docs/crim/uscioffitannin61808ind.pdf)
14
purpose of this paper is not to consider if Goldman was acting ethically and legally in
terms of its corporate governance, but rather to focus on the fact that Goldman
recognized the impairment of the financial instruments before others. How was that
possible? The recent case (the complaint) of Basis Yield Alpha Fund versus Goldman
Sachs & Co and affiliated companies can provide useful clues to solve this mystery.
Basis Yield Alpha Fund (BYAFM) brought an action against Goldman Sachs & Co. and
affiliated companies ("Goldman") for knowingly making materially false and misleading
statements and omissions in connection with the sale of a security issued by a
collateralized debt obligation (CDO) based upon subprime residential home mortgages,
known as Point Pleasant 2007-1, Ltd (Point Pleasant 24), and the entry into two credit
default swaps (CDS) that referenced AAA and AA rated securities from a similar CDO
known as Timberwolf 2007-1, Ltd (Timberwolf). Goldman began to market and sell the
Point Pleasant and Timberwolf securities in the first quarter of 2007. Within weeks of
BYAFM buying these securities, they rapidly declined in value, as the Defendants knew
and intended they would. Why Goldman could promptly recognize the decline of value in
those complex financial instruments backed by mortgages (Point Pleasant and Timberwolf)
and how it got the inside mark were presented as follows. 25
The market for securities based on subprime residential mortgages as it existed
during the timeframe relevant to this dispute was highly complex, opaque, and
concentrated. Only a few investment banks were significant issuers or traders in
this market, which was characterized by illiquidity and a paucity of publicly
available information. Goldman was a central participant in this market and was
intimately involved in all phases of it, including working closely with banks and
other lenders who made high-risk mortgage loans in the first instance, working with
syndicators in bundling mortgages into RMBS (residential mortgage backed
securities), creating and marketing both cash and synthetic CDOs, providing
information to rating agencies to secure ratings on the securities it was offering
and monitoring the performance of the securities and their constituent underlying
securities post-issuance. As a consequence, Goldman was one of a very small
group of market participants to have and acquire information about the current
value and outlook for RMBS and CDO offerings (para. 21).
Goldman and the other investment banks exercised substantial control over the
flow of information, including pricing information, about these RMBS and CDO
24
Actually, it was CDO squared security consisting of a portfolio of CDO assets or reference
obligations.
25 Basis Yield Alpha Fund (Master) v. Goldman, Index No: 652996/2011, Oct. 27, 2011.
15
securities. As a result, the investors in the RMBS and CDO securities relied heavily
on and reasonably expected the investment banks, such as Goldman, to provide
truthful and complete information about the RMBS and CDO securities and the
pricing and market for these securities (para. 22).
As the underwriter and sponsor of these securities, Goldman had far superior
knowledge to BYAFM about the quality, value, pricing and likely performance over
time of Point Pleasant and Timberwolf, as well as the criteria by which the
underlying and reference securities on which these offerings were based were
selected, information that was largely unavailable to BYAFM (para. 23).
On March 8, 2007, Daniel Sparks, the head of the Mortgage Department at
Goldman, gave, in an internal Goldman e-mail, a lengthy statement of his views on
RMBS. He referred to the Timberwolf deal, which at that point had not yet been
issued, as one of Goldman's "most risky" CDOs. He reconfirmed Goldman's
anticipation of a "dramatic credit environment downturn" and reiterated that
Goldman is "still net short” (para. 34).
Goldman knew that the resulting Point Pleasant and Timberwolf securities were of
much lower quality and value than represented. Goldman knew this because of its
own due diligence investigations and the due diligence investigations, performed
for Goldman's benefit but not for disclosure to customers Goldman was soliciting,
by outside firms like Clayton Holdings, Inc. ("Clayton"). In addition to arranging
CDOs, Goldman was also directly involved in acquiring or originating mortgage
loans and then assembling, creating, and marketing RMBS that were backed by
pools of these loans. As a direct result of this role, Goldman acquired a great deal
of non-public detailed information about the quality or lack thereof of the
mortgages that backed the RMBS. This information was highly material to
assessing whether the RMBS would either perform as expected, or instead would
fail to meet expectations and even go into default. Equally, this information was
highly material as to the expected performance and risk of CDOs constructed out
of these RMBS (Paras. 131-133).
Goldman, either directly or through a third-party due diligence firm, routinely
conducted due diligence review of the mortgage loan pools it bought from lenders
or third party brokers for use in its RMBS securitizations. Thus, Goldman, including
Sparks, had access to detailed non-public information concerning the true quality
of the loans collateralizing the RMBS securitizations it sponsored. Goldman
retained third-party due diligence providers such as Clayton to analyze the loans it
was considering placing in its securitizations. Throughout 2006, Goldman was
Clayton's largest client. For each quarter of 2006, and for the full year, Clayton
reviewed more loans for Goldman than for any other investment bank. Clayton told
the New York Attorney General "that starting in 2005, it saw a significant
deterioration of lending standards and a parallel jump in lending expectations." As
a key client of Clayton, Goldman had access to non-public reports and data by
Clayton showing this significant deterioration. Clayton's reports to Goldman were
confidential, and were not shared with purchasers of RMBS or CDOs underwritten
and sold by Goldman. Nor were the reports shared with BYAFM. Documents
released by Clayton confirm that Goldman was aware of the weakness in the loan
pool and in the underwriting standards of the originators it used in its RMBS
transactions (paras 138-142).
Although the above situation might have been an exceptional circumstance at the
16
beginning of the financial crisis, it indicates that useful information for estimating the
price of hard-to-value instruments would be limited to a small number of people in the
market, or it was unevenly distributed. However, in retrospect, the information that
Sparks (and Goldman) and Clayton had received at the beginning might have been close
to fair value of those securities. Yet, from an ethical viewpoint, it might be difficult to
claim that marks that stayed confidential in a couple of organizations and were not
widely distributed in the market would be fair value.
In reality, it is not uncommon for investment companies to use external professional
services for the valuation of some complex financial instruments. But, if several funds
received a feed from a common pricing service and did not exercise discretion adjusting
individual bond marks, then price clustering 26 would result (Cici et al. 2011, p. 214).
Some funds explicitly state that they mark bonds at mid-market prices (i.e., the average
of the bid and ask prices), whereas other funds explicitly state that they mark bonds at
bid prices (Cici et al. 2011, p. 215). It may be so because “different dealers experience
different customer flows and therefore could form different opinions about the underlying
value of any infrequently traded issue (Cici et al. 2011, p. 225).27 Fair value guidance
under SEC Accounting Series No. 118 is not extraordinarily restrictive, especially as it
permits funds to use a valuation within the range of bid and ask prices considered to best
represent value in the circumstances.
Furthermore, it is useful to take a look at the results of a survey conducted by one of
the Big 4 accounting firms (Deloitte 2010). Industry professionals representing 67 asset
managers completed a survey. Those asset managers advise more than 3,000 mutual
funds with assets under management exceeding $2.2 trillion. This survey result also
shows the disagreements are routine among asset managers.
One sign of the industry’s focus has been its use of the price challenge process.
Approximately 97% indicated that they have challenged the valuations by the
primary pricing vendor. Almost 34% have been issuing price challenges daily. The
rise of the price challenge process is not unusual given the volatility and
uncertainty in the investment marketplace and the general strengthening of fund
26
Cici et al. (2011) found evidence of clustering as well as a considerable spread prices across
funds.
27 Although the present study does not, Cici et al. (2011) explore if there is a possibility for
funds (or traders) to discretely manipulate marks to smooth their return.
17
policies and procedures. Fund groups are more successfully teaming within their
organizations, sometimes on a daily basis, to assess valuations provided by others
using traditional and newly developed internal tools. The challenge process and
questions asked have become routine. Portfolio managers and traders have
become more proactive in their assessment of the valuation process, and
additional relevant information is being considered (Deloitte 2010, p. 2).
The “price challenge” indicates disagreements between an asset manager and its
external pricing vendor. Requesting external vendors to offer pricing services28 in the
FVA process is now quite common. The survey results that explicitly indicate the
tendency are:






73% of survey participants indicated that they feel the pricing services provide a
more reliable valuation than brokers
97% indicated instances where they have challenged the valuations provided by
the primary pricing vendor
Pricing services are generally approved annually by 66% of fund groups
Brokers are more likely to be approved on an as needed basis (54%)
34% of survey participants indicated that they issue price challenges to pricing
vendors on a daily basis
When performing due diligence on pricing sources, 81% of survey respondents are
asking their pricing services and brokers whether prices reflect the most recent
transactions, an increase of 22% from the previous year’s survey
So far, the above discussion has more or less focused on the distributed aspect of
varied valuation practice in funds and external pricing vendors. In other words, it does not
fully consider the practice inside the organization. This is addressed in the next subsection.
3.2.3 Distributed FVA: Inside an Organization
Lépinay (2011, pp. 196-203) focuses on the creation and valuation of complex
financial instruments inside a French mega bank especially from the accounting
perspective. This study differs from those quoted above, in not explicitly mentioning
whether or not the bank used external pricing vendors. Rather, based on the character of a
financial instrument, the bank has three different configurations and each plays a different
role. They are back office, front office and accounting department. The back office has a
28
Mainly there are two types of pricing service: (a) Pricing services that use a proprietary
model to estimate a price and (b) Consensus pricing services (International Accounting
Standards Board Advisory Panel 2008, para. 64).
18
database, model and pricer 29, and for the valuation of the simple financial instruments the
accounting department does not need to rely on the front office’s traders’ knowledge.
Therefore, the back office directly gives information to the accounting department which
performs the transition from many calculations to one figure, and the latter is nothing but
an extension of the former (Lépinay 2011, p.197).
In contrast, for the valuation of new, complex and esoteric financial instruments, the
checks and controls of the back office are not enough. The accounting department relies
on the front office for the valuation (i.e. the trader) (Lépinay 2011, p.199). Sometimes, the
accounting department needs the back-office database in addition to the provision of
valuation-related information from the front office (Lépinay 2011, p.198).
Again, in the case of the valuation of Point Pleasant and Timberwolf depicted above,
information that Sparks had got from Clayton was not simply accepted by the other
members of Goldman. There was a valuation project where distributed actors in the
organization got together and discussed the appropriate valuation of CDOs. According to
the complaint30,
On May 11, 2007, Goldman senior executives, including Gary Cohn, Chief
Operating Officer and Co-President, and David Viniar, Chief Financial Officer, held a
lengthy meeting with Mortgage Department personnel, their risk controllers, and
others to develop a "Gameplan" for the CDO valuation project. The Gameplan
called for the Mortgage Department to use three different valuation methods to price
all of its remaining CDO warehouse assets and unsold securities from the
Goldman-originated CDOs, which included Timberwolf and Point Pleasant, then
being marketed to clients (para. 103).
May 20, 2007, in a late evening conference call in which Viniar, Sparks, Lehman,
and others participated. Sparks had reviewed and commented on a draft of the
presentation (the "Draft Presentation") in advance. The Draft Presentation identified
write-downs for CDO squared positions in the range of 111 to 221 million, and stated
that these write-downs "are driven by Mezz AAA and AA tranches of Timberwolf and
Point Pleasant CDO2 [CDO squared] structures” (paras. 104-105).
A subsequent internal Goldman analysis of prices for Timberwolf was conducted in
preparation for the May 20 management meeting. This analysis showed two sets of
29
Pricers are pieces of software that indicate financial products’ values and help traders run
another decision making machine, the risk-analysis software. They work almost automatically
and produce a price on demand. Without the pricer, the value of the securities in a portfolio
would be indeterminate (Lépinay 2011, p. 48 and 75). It should be noted that the back office
and the front office have its own pricer.
30 Basis Yield Alpha Fund (Master) v. Goldman, Index No: 652996/2011, Oct. 27, 2011.
19
prices for the Timberwolf tranches that were ultimately purchased by BYAFM, one
set based on the CDO marks, the other on RMBS marks. The prices developed
through this analysis were (para. 114):
Timberwolf AAA
Timberwolf AA
Price based on CDO
Marks
66
38
Price based
RMBS Marks
24
15
on
As this quotation indicates, although Sparks seemed to have more knowledge than others
with regard to valuation, his valuation was not simply accepted. Even among the highly
sophisticated Goldman professionals there were different ranges of marks for valuation.
The Deloitte’s survey results (Deloitte 2010) also indicated that the pricing
committee 31 tends to be independent of senior board members in the FVA process.
Furthermore,


58% of survey participants indicated that the Board of Trustees has created a
separate Fair Valuation Committee
48% of survey participants indicated that only non-interested trustees serve on the
Board’s Fair Valuation Committee. 39% indicated that the involvement is the same
amongst interested and non-interested trustees.
3.2.4 Summary
From the above considerations of practice in different types of funds, two points are
notable. First, marks used for pricing some complex and thinly-traded financial instruments
are markedly different among valuers and this could create conflicts among them. Thus,
many financial institutions or funds are using valuations by external professionals.
However, the pricing challenges still happen occasionally in some funds. Second, the
process of deciding on one value or number especially for financial accounting purposes
can be distributed among different types of organizations or individuals. As depicted in
Figure 2, those agents may be located inside as well outside the organization whose
purpose is trading sophisticated financial instruments.
31
Most funds have valuation committees, but practices vary widely as to their precise role and
composition and the frequency with which they meet (ICI et al. 2005).
20
Inside the organization
Back Office
Pricer/
Data
base
Accounting
Department/
Group of
Officers/Fair
Valuation
Committee
Front Office
Pricer
Investors
Outside the
organization
Other
Organizations
Pricing
Vendors
Pricer/
Data
base
Pricer/
Data
base
Figure 2 Possible Distribution of FVA
This figure is based on Figure 22 of Lépinay (2011).
4.
An Alternative Perspective on FVA
According to the above analyses, the valuation of some complex financial instruments
in the financial market may be different in every organization, and it might be in an extreme
case that valuations are unevenly distributed to limited numbers of organizations. They
might be distributed first to limited prestigious companies first and then to others because
prestige and fame could be important dimensions in markets (Lépinay 2011, p. 250).
Furthermore, in investment companies the valuation of some financial instruments can not
be decided by traders, a back office, or an accounting department in isolation, but must be
done in conjunction with other departments or external organizations. Therefore, because
of the distributed knowledge of FVA, the final figure comes down to a dance of those
agencies (Pickering and Guzik 2008). Given the nature of this decision making, insights
from the studies of distributed cognition could be useful to interpret the process.
4.1 Insights from the Study of Distributive Cognition
One of the most prominent studies on distributed cognition is Hutchins’ work (Hutchins
21
1995a and 1995b) which is based on extended studies of ship navigation and cockpit of a
commercial airline. His work mainly focuses on the fact that “human cognition is always
situated in a sociocultural world and cannot be unaffected by it (1995a, xiii).” The concept
of distributed cognition has attracted attention in several fields, ranging from law (e.g. jury
decision making) and sociology (e.g. information processing in organizations) to computer
science (e.g. GRID computing or medical informatics) and the philosophy of science (e.g.
expert panels) (List 2008, p. 285). For this paper, this perspective provides three useful
points for considering FVA.
The first point relates to the division of labor. According to Hutchins (1995a, p. 176),
“All division of labor, whether the labor is physical or cognitive in nature, requires
distributed cognition in order to coordinate the activities of the participants.” Callon and
Muniesa (2005, p. 1245) also state “Economic calculation is not an anthropological fiction,
precisely because it is not a purely human mechanical and mental competence; it is
distributed among human actors and mental devices.” So, “rather than simply assuming
that all cognition is restricted to individuals, we are invited to think of some actual cognition
as being distributed among several individuals” (Giere 2007, p. 314). By doing so, the gap
that exists between models of a general-purpose cognitive architecture and explanations
of human performance in complex, real-word situations can be filled (Hazlehurst et al.
2008, p. 230). With this view, “a cognitive science that is neither mentalistic (remaining
agnostic on the issue of representation “in the head”) nor behavioristic (remaining
committed to the analysis of information processing and the transformation of
representations “inside the cognitive system” (Hutchins 1995a, p. 129)). This perspective
of distributed cognition can be an alternative to an individual-centered model of cognition
(Hazlehurst et al. 2008, p. 227). An essential condition for distributed cognition in a group
is the presence of an organizational structure that allows the group to produce collective
judgments.32 Based on the analysis in the previous sections, FVA related decisions in
investment companies seem to be based on the collective judgment which consists of
diversified knowledge distributed to inside and outside the organization.
32
Collective judgments are outputs of representational content of a distributed cognitive
system (List 2008, p. 288).
22
The second point raised by the perspective of distributed cognition concerns the
extent to which the cognition is distributed. In the framework, it is not impracticable to
assume that cognition can be distributed not only to human beings but also to tools or
artifacts. Berdone and Secchi (2009, p. 190) note that external resources, such as artifacts,
tools, and objects, shape human cognition. MacKenzie (2009) assumes that cognition and
calculation can be distributed not only to human beings but also to artefacts, such as
technical system. According to Hazlehurst et al. (2008, p. 228), distributed cognition treats
the activity system, rather than the individual, as the unit of cognitive analysis. An activity
system comprises a group of human actors, their tools and environment, and is organized
by a particular history of goal-directed action and interaction. So, in addition to human
actors, the configuration of information-bearing tools, such as a database or a pricer, can
play some functional role using this framework.
The third noticeable point with the theory of distributed cognition is that our cognitive
systems are not passively affected by external resources. Rather, human beings actively
overcome their internal limitations by (1) externalizing and disembodying thoughts, ideas,
and solutions, and then (2) re-projecting internally what occurs outside in the external
invented structure to find new ways of thinking (Berdone and Secchi 2009, p. 193). During
the externalization process, individuals create something that exists without their brains ,
and the invented structure, such as an artifact or a tool, can eventually be useful to other
individuals (Berdone and Secchi 2009, p. 193). Consequently, it constitutes the basis for
social interactions. In short, the distribution of cognition is mobilized by externalization, and
it tends to be mimetic (Berdone and Secchi 2009, p. 193). The externally distributed
cognitive framework is then repeatedly and creatively projected to solve cognitive tasks
(Berdone and Secchi 2009, p. 193). Thus, the manner in which individuals are able to
interact with externally invented structures and re-project them according to their own
knowledge is a contentious issue.
In summary, these three points offer insights to current FVA practice. First, limited
human capacity, which makes distributed cognition significant, could be the basis for
understanding the distribution of FVA practice. Second, the importance of material
databases and pricers in FVA is compatible with the ontological stance of the distributive
23
cognition perspective. Third, the emphasis on the externalization process in the framework
of distributive cognition gives a dynamic perspective on interactions between artifacts and
human beings in the process of FVA. It is certain that those processes are mainly
determined by human beings, which is the point discussed further in the next subsection.
4.2 Interactions of Human Beings and FVA
In terms of Giere’s (2005) perspectivism, practice of FVA in investment companies
may involve the operation of distributed systems and incorporate different pricers and
databases. In the actual imperfect market conditions, FVA invariably involves such an
operation. Thus, “not only is the production of company accounts beyond the powers of an
individual, but limited “hardness” the resultant numbers possess is intrinsically bound up
with the involvement of multiple human beings in roles that are structured in part
technologically” (MacKenzie 2009, p. 17). Furthermore, according to Schmidt (2009, p.
274), the final figure of FVA tends to be determined by the individual parties’ negotiation
skills and powers. For example, the process of price challenging is undoubtedly a type of
negotiation. For that type of negotiation, various factors are critical. For example, “Various
cultural phenomena―such as morals and related norms and values, religious and other
beliefs, ideas, ideologies and ideals, traditions and customs, ethnicity and nationality, and
so on―all have definite influences on market processes and outcomes, particularly on
both product prices and factor prices (incomes)” (Zafirovski 2000, p. 289).
But, the way that this knowledge can be externalized and re-projected is not
thoroughly investigated in this paper.33 So far, this paper has discussed the process of the
distribution: imitative and creative behaviors. Here, it is beneficial to provide some
normative comments on this aspect. To make such tentative comments objectively, recent
evidence in cognitive science is simply quoted. The normative suggestion can throw a
unique light on the issue of objective/subjective disputes concerning the fairness of FVA.
In the study by McCabe et al. (2001), the subjects played a trust and reciprocity game
(iterated Prisoner’s Dilemma) against a human counterpart or a computer. In the subjects
For example, List (2008) focuses on a group’s aggregation procedure within the distributed
cognitive system.
33
24
who co-operated there was more activity in the MPFC (medial prefrontal cortex) when
playing with a person than with a computer. MPFC is one of the brain regions consistently
activated when subjects perform tasks in which they have to think about the mental states
of others (Firth and Singer 2008, p. 3878). This implies that a person i s inherently fairer
when he/she interacts with another person rather than with machines. Interactions of
internal and external professionals through their distributed knowledge in current FVA
entail some degree of human connection. The valuation is not always finalized with models,
databases, and pricers. Even if a specific group of individuals in an investment company
has leading-edge pricers or highly-efficient machines, a fairer valuation is achieved by
interactions among individuals. This hypothetical implication is in line with an assertion of
social study of science underpinned by the distributed cognition perspective that “a task of
valuation is often performed not by a single human being unaided but by multiple human
beings” (MacKenzie 2009, p. 16). For example, human beings have a special tendency for
reciprocity 34 or docility 35 that models, concepts, and data do not possess. “It is the
humans, and only the humans, that provide intentional, cognitive agency to scientific
distributed cognitive systems” (Giere 2007, p. 319). Human beings have a capacity for
collective intentionality that makes them share intentional states such as beliefs, desires,
and intentions (Searle 1995, p. 23). This concept can be partially useful in understanding
that a human actor implicitly or explicitly assumes that others’ cognition is similarly
distributed to human beings and artifacts.36
This implies that the fair value of hard-to-value financial instruments can be fairer if
collective knowledge of the instruments is distributed to market participants and widely
shared. However, current valuation practices in investment companies are based on quite
sophisticated models and software. Even in that situation, how human beings with unique
reciprocity and docility interact with other individuals and materials, such as databases and
34
See the results of ultimatum game experiments in Firth and Singer (2008).
This relates to the fact that we support our limited decision-making capabilities by receiving
inputs, perceptions, data, and so on from the social environment (Berdone and Secchi 2009, p.
194).
36 However, Giere (2007, p. 319) seems not to agree with this point by stating, “To understand
how the members of the groups collectively make the system work, it is not necessary, and I
think definitely unhelpful, to introduce the concept of a super, or collective, agent.”
35
25
models, requires further elaboration. Callon (2004, p. 7) states that “ The more we
recognize that non-humans have an active social role, the more we enrich human nature. A
consequence of the rehabilitation of non-humans is thus the rehabilitation of humans.”
5.
Conclusion
Many proponents of FVA implicitly assume that financial markets are efficient. “Fair
value valuations are supposed to both reflect and reinforce that efficiency, the latter by
better diffusion of real time information on objective assets and liability valuations ”
(McSweeney 2009, p. 837). However, there is no conclusive evidence that FVA should
totally replace other valuation methods such as HCA, and the mixed-attribute valuation
model has been accepted in accounting regulation for a long time.
This study tentatively revealed, on the basis of some glimpses of practice of FVA for
financial instruments in different types of organizations, that the process as well as
technical knowledge of FVA could be distributed among databases, models, software as
well as specialist individuals who possess their own knowledge. This point can be
explained from the perspective of distributed cognition. Although this study assumes that
the knowledge could be distributed not only among different individuals but also among
other models, databases, and software, it suggests that more interactions among human
beings are necessary for fairer valuation. “We often assume that a valuation of technology
is accepted in practice because of technical superiority, but it is also a sociological truth
that such technologies will seem superior if they are widely accepted and if they can
appeal to deeply held cultural values” (Power 2010, p. 209).
In particular, considering modern complex financial markets, professionalism cannot
promise accountants personal autonomy with respect to their work. Such autonomy would
be both impossible and counterproductive in the context of the collaborative and
interpretative relationships through which accounting knowledge is produced (Gill 2009, p.
108). In sum, the distribution of the process of FVA may be inevitable, but the distribution
underpinned by interactions of human beings beyond the boundaries of inside and outside
the organizations can be also important. For that, accounting standards setters could
consider this aspect of distribution of FVA and adjust the trajectory of standards for FVA.
26
With these assertions as a basis, a future study could explore how the knowledge and
the processes of FVA distributed to individuals or organizations (such as traders in the front
office, staff in the back office and the accounting department, professionals in external
vendors, auditors in accounting firms 37) interact.
References
AAA (American Accounting Association) Financial Accounting Standards Committee (2005),
“Response to the FASB’s Exposure Draft on Fair Value Measurements,” Accounting
Horizons, Vol. 19, No. 3, pp. 187-196.
Ake, J. N. and Hays, M. J. (2007), “Regulatory and Accounting Standards on the Valuation of
Portfolio Securities by Investment Companies,” Wall Street Lawyer, Vol. 11, No. 1, pp. 1-8.
Berdone, E. and Secchi, D. (2009), “Distributed Cognition: A Research for Management,”
Current Topics in Management, Vol. 14, pp. 183-207.
Bignon, V., Biondi, Y. and Ragot, X. (2009), “An Economic Analysis of Fair Value,” Prism N o15
(August 2009), Available at SSRN: http://ssrn.com/abstract=1474228.
Bougen, P. D. and Young, J. J. (2011), “Fair Value Accounting: Simulacra and Simulation,”
Critical Perspectives on Accounting, forthcoming.
Brown-Hruska, S. and Satwah, S. (2009), “Financial Disclosure and SFAS 157: Seeking
Transparency in a Perfect Storm”, Capital Markets Law Journal, Vol. 4, No. 2, pp. 133-154.
Callon, M. (2004), “The Role of Hybrid Communities and Socio-technical Arrangements in the
Participatory Design,” Journal of the Center for Information Studies, Vol. 5, No. 3, pp. 3-10.
Callon, M. and Muniesa, F. (2005), “Economic Markets as Calculative Collective Devices,”
Organization Studies, Vol. 26, No. 8, pp. 1229-1250.
Campbell, R. L., Owens-Jackson, L. A. and Robinson, D. R. (2008), “Fair Value Accounting
from Theory to Practice: Implementation of SFAS No. 157 will be Tumultuous, Painful, and
Require Significant Adjustment for U.S. Accounting Professionals,” Strategic Finance, Vol.
90, No. 1, pp. 31-37.
Cici, G., S. and Merrick Jr., J. J. (2011), “Missing the Market? Dispersion in Corporate Bond
Valuations across Mutual Funds,” Journal of Financial Economics, Vol. 101, No. 1, pp.
206-226.
Cohan, W. D. (2011), Money and Power, Penguin Books: New York.
Deloitte (2010), Springing Forward: The Eighth Annual Fair Value Pricing Survey , Executive
Summary and Key Trends and Findings.
Dore R. (2009), “Financialization of the Global Economy,” Industrial and Corporate Change,
Vol.17, No. 6, pp. 1097–1112.
Epstein, R. (ed.) (2005), Financialization and the World Economy, Edward Elgar: Aldershot.
Epstein, R. and Henderson, M. T. (2011), “Do Accounting Rules Matter? The Dangerous Allure
of Mark to Market,” Journal of Corporation Law, Vol. 36, No. 3, pp. 513-549.
Firth, C. D. and Singer, T. (2008), “The Role of Social Cognition in Decision Making,”
Philosophical Transactions of the Royal Society B: Biological Sciences , Vol. 363, No. 1511,
pp. 3875-3886.
Financial Accounting Standards Board (FASB) (2008), Statement of Financial Accounting
Standard No. 157 Fair Value Measurement (Original Pronouncements as Amended).
Fitzsimons, A. P., Satenstein, J. L. and Silliman, B. R. (2010), “Fair value Audit Guidance of
Public and Non-Public Companies in Response to the Credit Crisis,” Review of Business,
According to Goncharov et al. (2011), “audit fees are decreasing in the firm’s exposure to
assets reported at fair value, and increasing in the firm ’s exposure to more difficult-to-measure
(i.e., levels 2 and 3) fair values.
37
27
Vol. 30, No. 2, pp. 40-49.
Galligan D. (2007), “Fair Value Measurement for Corporate Entities, Insurance Companies and
Retail Banks, An Investment Banker’s Perspective,” in Walton, P. (Ed), The Routledge
Companion to Fair Value and Financial Reporting, pp. 265-277.
Georgiou, G. and Jack, L. (2011), “In Pursuit of Legitimacy: A History behind Fair Value
Accounting,” British Accounting Review, Vol. 43, No. 4, pp. 311-323.
Giere, R. N. (2005), Scientific Perspectivism, the University of Chicago Press: Chicago.
Giere, R. N. (2007), “Distributed Cognition without Distributed Knowing,” Social Epistemology,
Vol. 21, No. 3, pp. 313-320.
Gill, M. (2009), Accountants’ Truth, Oxford University Press: Oxford.
Goncharov, I., Riedl, E. J. and Sellhorn, T. (2011), “Fair Value and Audit Fees,” (May 30, 2011),
Available at SSRN: http://ssrn.com/abstract=1855628.
Hazlehurst, B., Gorman, P. N. and McMullen, C. K. (2008), “Distributed Cognition: An
Alternative Model of Cognition for Medical Informatics,” International Journal of Medical
Informatics, Vol. 77, No. 4, pp. 226-234.
Hitz, J. (2007), “The Decision Usefulness of Fair Value Accounting – A Theoretical
Perspective,” European Accounting Review, Vol. 16, No. 2, pp. 323-362.
Huault, I. and Montagner, R. (2009), “Market Shaping as an Answer to Ambiguities: The Case
of Credit Derivatives,” Organization Studies, Vol. 30, No. 5, pp. 549-575.
Hutchins, E. (1995a), Cognition in the Wild, MIT Press: Cambridge.
Hutchins, E. (1995b), “How a Cockpit Remembers its Speed,” Cognitive Science: A
Multidisciplinary Journal, Vol. 19, No. 3, pp. 265-288.
International Accounting Standards Board Advisory Panel (2008), Measuring and Disclosing
the Fair Value of Financial Instruments in Markets that are No Longer Active, October
2008.
Investment Company Institute (ICI) Independent Directors Council and ICI Mutual I nsurance
Company (2005), Fair Valuation Series: An Introduction to Fair Valuation.
Investment Company Institute (ICI) Independent Directors Council and ICI Mutual Insurance
Company (2006), Fair Valuation Series: The Role of the Board.
International Monetary Fund (IMF) (2008), Global Financial Stability Report: Financial Stress
and Deleveraging Macrofinancial Implications and Policy, October 2008.
Jacquemard, R. (2007), “Fair Value and the Auditor,” in Walton, P. (Ed), The Routledge
Companion to Fair Value and Financial Reporting, pp. 278-299.
Kaplan, R. S. (2011), “Accounting Scholarship that Advances Professional Knowledge and
Practice,” The Accounting Review, Vol. 86, pp. 367-383.
Koga, C. (2001), “Financial Instruments and Finance-Oriented Accounting [in Japanese],”
Kokumin Keizai Zasshi, Vol. 183, No. 5, pp. 31-43.
KPMG (2011), IFRS Illustrative Financial Statements: Investment Funds, December 2011.
Leung, D. (2011), Inside Accounting-The Sociology of Financial Reporting and Auditing, Gower.
Lépinay, V. A. (2011), Code of Finance, Princeton University Press: Princeton.
List, C. (2008), “Distributed Cognition: a Perspective from Social Choice Theory,” in Albert, Max,
A., Dieter, S. and Stefan, V. (eds.), Scientific Competition: Theory and Policy. Conferences
on New Political Economy, Mohr Siebeck, Tübingen: Germany, pp. 285-308.
MacKenzie, D. (2009), The Material Markets-How Economic Agents are Constructed, Oxford
University Press: Oxford.
Martin, R. D., Rich, J. S. and Wilks, T. J. (2006), “Auditing Fair Value Measurements: A
Synthesis of Relevant Research,” Accounting Horizons, Vol. 20, No. 3, pp. 287-303.
McCabe, K., Houser, D., Ryan, L., Smith, V. and Trouard, T. (2001), “A Functional Imaging
Study of Cooperation in Two-person Reciprocal Exchange,” Proceedings of the National
Academy of Sciences of USA, Vol. 98, No. 11, pp. 832–835.
McSweeney, B. (2009), “The Roles of Financial Asset Market Failure Denial and the Economic
Crisis: Reflections on Accounting and Financial Theories and Practices, ” Accounting,
Organizations and Society, Vol. 34, Nos. 6-7, pp. 834-848.
Millo, Y. and MacKenzie, D. (2009), “The Usefulness of Inaccurate Models: Towards an
Understanding of the Emergence of Financial Risk Management,” Accounting,
Organizations and Society, Vol. 34, No. 5, pp. 638-653.
Obinata, T. (2011), “Misuses of Fair Value and Financial Crisis [in Japanese],” CARF Working
28
Paper (J-073), Centre for Advanced Research in Finance, University of Tokyo.
Penman, S. (2008), “Principles for the Application of Fair Value Accounting,” White Paper, No.
2, Centre for Excellence in Accounting and Security Analysis (Columbia Business School).
Penman, S. (2010), Accounting for Value, Columbia Business School Publishing: New York.
Pickering, A. and Guzik, K. (eds.) (2008), The Mangle in Practice: Science, Society, and
Becoming, Duke University Press: Durham.
Pickering, A. (2009), “The Politics of Theory-Producing another World, with Some Thoughts on
Latour,” Journal of Cultural Economy, Vol. 2, Nos. 1-2, pp. 197-212.
Power, M. (2010), “Fair Value Accounting, Financial Economics and the Transformation of
Reliability,” Accounting and Business Research, Vol. 40, No. 3, pp. 197-210.
PricewaterhouseCoopers (PwC) (2011), Illustrative IFRS Financial Statements 2011 ―
Investment Funds.
Ramana, K. and Watts, R. L. (2011), “Evidences on the Use of Unverifiable Estimates in
Required Goodwill Impairment,” Working Paper (Harvard Business School), 09-106.
Ryan, S. G. (2008), “Fair Value Accounting: Understanding the Issues Raised by the Credit
Crunch,” White Paper, the Council of Institutional Investors.
Scarlata, J. and Spackman, C. (2009), “Credit Derivatives: Systematic Risk and Policy
Options?” IMF Working Paper (WP/09/254), International Monetary Fund (November).
Schmidt, M. (2009), “Fair Value: Your Value or Mine? An Observation on the Ambiguity of the
Fair Value Notion Illustrated by the Credit Crunch,” Accounting in Europe, vol. 6, No. 2, pp.
271-282.
Singer, T. and Fehr, E. (2005), “The Neuroeconomics of Mind Reading and Empathy,” American
Economic Review, Vol. 95, No. 2, pp. 340-345.
TheCityUK (2011), Financial Market Series ― Fund Management, October 2011.
The Financial Crisis Inquiry Commission (2011), The Financial Crisis Inquiry Report, Final
Report of the National Commission on the Causes of the Financial and Economic Crisis in
the United States, January 2011.
Young J. J. and Williams, P. F. (2010), “Sorting and Comparing: Standard-setting and ‘Ethical’
Categories,” Critical Perspectives on Accounting, Vol. 21, No. 6, pp. 509-521.
Zafirovski, M. Z. (2000), “An Alternative Sociological Perspective on Economic Value: Price
Formation as a Social Process,” International Journal of Politics, Culture, and Society, Vol.
14, No. 2, pp. 265-295.
29